A rate-limiting step in the evaluation of molecularly targeted cancer therapeutics is the ability to match the right inhibitor with the right patient. Success requires molecularly profiling of tumor tissue from patients and the discovery of biomarkers that predict sensitivity to specific agents. This project addresses this question in prostate cancer with the focus on inhibitors of the PI3-kinase (PI3K) pathway. The project builds on progress by Drs. Sawyers and Wu during the first 5 years of funding studying mTOR inhibitors and capitalizes on their expertise in developing genetically engineered prostate mouse models and using these models to evaluate signaling pathway inhibitors. Because the concepts in this project are clearly ripe for clinical translation, we view Dr. Sawyers' recent relocation to Memorial Sloan-Kettering Cancer Center (MSKCC) as an asset, providing access to an outstanding clinical prostate cancer group directed by Drs. Howard Scher and Peter Scardino, that will complement the clinical program at UCLA. The project will use three well-characterized genetically engineered prostate cancer models (PTEN, MYC, AKT) to: (i) define and measure pathway- specific, prostate cancer biomarker signatures in tumor tissue and serum using mouse models and human patients and (ii) evaluate PIS kinase pathway inhibitor therapy alone and in combination using hormone- dependent and hormone refractory prostate cancer mouse models. These findings will guide the design and execution of neoadjuvant clinical trials of PI3 kinase pathway inhibitors in prostate cancer, collaboratively at MSKCC and UCLA, and assess the utility of circulating tumor cells (CTCs) as oncogenomic readouts in prostate cancer patients.